Human Ecology, Vol. 15, No.

2, 1987

The Agroecology of Corn Production

in Tlaxcala, Mexico
Miguel A . Altieri 1 and Javier Trujillo 1

The primary components of Tlaxcalan corn agriculture are described, including

cropping patterns employed, resource management strategies, and interactions of
human and biological factors. Tlaxcalan farmers grow corn in an array of polyculture and agroforestry designs that result in a series of ecological processes
important for insect pest and soil fertility management. Measurements derived from a few selected fields show that trees integrated into cropping systems modify the aerial and soil environment of associated understory corn
plants, influencing their growth and yields. With decreasing distance from
trees, surface concentrations of most soil nutrients increase. Certain tree species affect corn yields more than others. Arthropod abundance also varies
depending on their degree of association with one or more o f the vegetational components of the system. Densities of predators and the corn pest Macrodactylus sp. depend greatly on the presence and phenology o f adjacent
alfalfa strips. Although the data were derived from nonreplicated fields, they
nevertheless point out some important trends, information that can be used
to design new crop associations that will achieve sustained soil fertility and
low pest potentials.
KEY WORDS: corn agroecosystem; agroecology; polycultures; agroforestry; traditional farming.

INTRODUCTION

Traditional farming systems in the Third World have emerged over centuries of cultural and biological evolution and represent accumulated ex-

1Division of Biological Control, University of California, Berkeley, California 94720.

189
0300-7839/87/0600-0189505.00/09 1987Plenum PublishingCorporation

190

Altieri and Trujillo

periences of interacting with the environment by farmers without access to

external inputs, capital, or scientific knowledge (Chang, 1977; Wilken, 1977;
Egger, 1981). Such experience has guided farmers in many areas to develop
sustainable agroecosystems, managed with locally available resources and
with human/animal energy (Harwood, 1979; Klee, 1980). These
agroecosystems, based on the cultivation of a diversity of crops in time and
space, have allowed traditional farmers to maximize harvest security with
limited resources and space (Clawson, 1985).
As change occurs in the face of inevitable agricultural modernization,
genetically variable indigeneous varieties of crops are being replaced by new
high-yielding ones (Harlan, 1976), and knowledge of the traditional cropping patterns and management practices and the ecological rationale behind
them are being gradually lost (Chambers, 1983). Modern agricultural development, characterized by broad-scale technological recommendations, has largely ignored the environmental, cultural, and socioeconomic heterogeneity
typical of traditional agriculture (Conway, 1985). Large-scale promotion of
uniform crop varieties, technologies, and farming systems, largely inaccessible to the great majority of peasants, has created an inevitable mismatching of agricultural development and the needs and potentials of local people
(deJanvry, 1981).
A more appropriate agricultural strategy requires a holistic approach,
sensitive to the complexities of agroecological and socioeconomic processes,
which can lead to the design of farming systems and techniques tailored to
the needs of specific peasant groups and regional agroecosystems (Altieri and
Anderson, 1986). A challenge of this nature requires an understanding of
the features of traditional agriculture such as the ability to bear risk, labor
constraints, diet preferences, and the spatial and temporal dynamics of symbiotic crop mixtures.
Although assessments of traditional "know-how" and anthropological,
geographical, and ethnobotanical studies of traditional farming patterns in
the Third World are numerous (Bye, 1981; Denevan, Treacy, Alcorn, Padoch,
Denslow, and Flores Paltan, 1984; Klee, 1980; Brokenshaw, Warren, and
Werner, 1980; Posey, 1983; Vermeer, 1979; Weinstock, 1983; Alcorn, 1981,
1984; Brush, Carney, and Huaman, 1981; Alverson, 1984; Grossman, 1984;
Gliessman, Garcia, and Amador, 1981; Marten, 1986). They have mainly
focused on the patterns of behavior followed by an ethnographic unit in the
realm of agricultural technology, which result in unique sets of land utilization in time and space, seasonal distribution of labor, nutrition, and other
needs. Rarely have studies examined the ecological mechanisms that underlie the sustainability of these agroecosystems. Similarly, agronomic and economic evaluations of traditional agroecosystems are routinely conducted
during Farming Systems Research (FSR) projects (Byerlee, Collinson,

Corn Production in Tlaxcala

191

Perrin, Winkelman, and Biggs, 1980; Harwood, 1979; Shaner, Phillip, and
Schenhl, 1982; Zandstra, Price, Litsinger, and Morris, 1981). Although FSR
surveys analyze the natural and socioeconomic circumstances affecting small
farms and attempt to involve the farmers themselves in all stages of the
research and development process, their emphasis on higher yield inevitably
still leads to the continual recommendation of high input technology (Oasa,
1985), with no apparent understanding of the ecological processes involved.
Although greatly modified, most traditional agroecosystems still utilize
minimal inputs, lack continuous disturbances, and exhibit complex interactions
among and between people, crops, soils, animals, etc., which makes them unique
settings from which to extract ecological principles for the design of new, improved, and locally adaptable sustainable agroecosystems (Altieri, 1983). Therefore, it is crucial to ecologically analyze traditional farming systems, not only
to evaluate their properties of stability, equitability, and sustainability, but also
to determine which elements of resource use and conservation should be retained during the course of agricultural modernization.
With this in mind we studied the agriculture of the state of Tlaxcala,
Mexico (Benitez, 1953; Wilken, 1977). Tlaxcalan farmers have continuously
faced problems of low quality natural resources, low soil fertility, and extreme climatic limitations. They have managed to survive in this fragile environment through a process of ecological modification and adaptation based
on diversified farming, genetic diversity maintenance, and unique land and
water management practices.
As part of a long-term research project begun in 1982, ecology of cornbased traditional farming systems has been studied in order to derive agroecological principles that may serve to guide future agricultural development
in the region. Corn production is the basis of Tlaxcalan agriculture, and corn
fields are diversified at both the species and genetic level. Corn is grown mixed
with annual and perennial plants in polycultural and/or agroforestry patterns, and also in genetically complex fields that maintain numerous genotypes over time and space. These systems provide a unique opportunity to
examine the ecological consequences of crop diversity patterns on conservation of soil and water resources, pest regulation, stabilization of production,
and on the long-term sustainability of the agroecosystem as a whole. We
qualitatively describe here the general structure and management of traditional corn production systems. With data derived from a few selected fields,
we also attempt to illustrate the ecological dynamics of the agroecosystems,
with some quantitative descriptions of the influence of vegetational diversity on plant-soil relationships and on the population levels of insect pests and
associated natural enemies. We focused on these processes because they seem
to be important indicators of the dynamics of most mixed farming systems
(Altieri, 1983, Altieri and Letourneau, 1982).

192

Altieri and Trujiilo

Fig. 1. A descriptive map of Tlaxcala, Mexico.

R E G I O N A L S E T T I N G OF T H E STUDY
Tlaxcala is located in east central Mexico in the highlands of the
Neovolcanic axis on the Mesa Central (Fig. 1). The state, situated above 2300
m elevation, is bisected from the northwest to the southeast by a wide flat
alluvial corridor bordered by low-lying hills and mesas. The climate of Tlaxcala is characteristically temperate (subhumid). Rainfall occurs from M a y October, ranging from 500 mm per year east of Huamantla to 800-1000 mm
in the southwest. Rainfall variations in the mid-summer months can lead to
extended droughts. Average monthly temperatures fluctuate within a narrow range, with January being the coldest month (0-9~ and April or May
generally the warmest (19-27~
Frosts are frequent and are particularly
threatening from November-February. In about 40~ of the state, it hails
an average of 2-4 days per year (Anonymous, 1981).
Tlaxcalan soils are generally sandy and highly drained, though some
soils are gravelly or rocky. Depth varies from 10 cm in the Lithosols of the
west and north central regions to deep Fluvisols on the plains of Huamantla. Most soils are infertile due to lack of organic matter. Soil pH tends to
be slightly acid (Anonymous, 1981). Much of the sloping land under agriculture is affected by erosion in different degrees. Tlaxcala has only two superficial waterways of importance: the Rio Zahuapan and Rio Atoyac.
subterraneal water is the principal source of supply for all types of use.

Corn Production in Tlaxcala

193

More than 70% of the total land area of the state (nearly 2900 km 2)
is dedicated to the production of corn, barley, wheat, beans, fava beans.
potatoes, and some vegetables (Anonymous, 1984). Less than 4% is irrigated. Most of the land (71%) is government owned as part of the land reform
program and farmed by ejiditarios (Anonymous, 1984). Average farm size
in the state is 4.83 ha, where only a few families are able to produce sufficient harvests to both feed themselves and to sell in local markets for additional income. Most must look for off-farm work, many as day laborers
O'ornaleros) on other farms, or as workers in nearby cities such as Puebla
or Mexico, D.F. during the off-season. Farm labor is generally supplied by
all family members. Many farmers do exchange work with neighbors or other
extended family, especially during peak demand periods such as harvesting.
Rarely is labor hired except by some of the largest landowners.

METHODS

The information presented here was obtained through 3 years of field

surveys, biological data collections, reviews of published materials, and interviews with farmers, agricultural technicians, and extensionists. Fifty farmers were interviewed throughout the state to gather information descriptive
of farm designs, crop patterns, use of local resources, and farmers' practices, as well as to understand why, in light of their particular circumstances,
farmers adopt such cropping systems and production methods. The interviews were also designed to assess the factors affecting farmers' decisions
with respect to the use of crop technologies, and how they avoid risk and
adapt to natural and socioeconomic constraints. Direct interviews and observations provided detailed information about crop management practices,
i.e., land preparation, planting, weeding, pest control, fertilization, harvest,
etc., their timing, and amount of inputs utilized in farm operations (Byerlee
et al., 1980).
Field surveys included the general description of the landscape, identification of existing crop patterns, description of cropping systems' determinants, and the actual and potential interactions between the various
biological components in the various polycultures and agroforestry systems
(Shaner et al., 1982). Field measurements were conducted in a few selected
fields to quantify the effects of trees on chemical and physical soild properties and corn yields, and to elucidate the population trends of pest insects
and associated natural enemies, and the resulting degree of pest damage in
corn systems grown under various levels of vegetational diversity.
Since all fields utilized were comprised of simple, nonreplicated experimental units, an inherent complication associated with research in farm-

194

Altieri and Trujillo

ers' fields, inferential statistics were not applied to the data. However,
biological knowledge and intuition were used to appraise whether the observed differences in measured parameters between fields were due to the
effects of differential diversification.
Tree Influences on Soil Properties and Corn Yields

In 1983, four rainfed cornfields in southeastern Tlaxcala with scattered

capulin (Prunus capuli) and sabino (Juniperusdeppeana) trees were selected
for the study (Farrell, 1984).
Four similar-size (8-10 m) trees of each species were selected in each
field. Five zones were chosen around each tree representing various areas
of tree influence within which sampling was carried out. These included the
canopy zone (C), morning shade zone (S-W), afternoon shade zone (S-E),
root zone (R), and zone of no tree influence (NI; Zinke, 1962).
Surface soil samples to a depth of 15 cm were collected at 2, 6, 10, 14,
18, and 22 m from the base of each tree along four randomly chosen transects. Soil properties measured were: total nitrogen (by macropkjeldahl),
water soluble phosphorus (determined colorimetrically), total carbon (by dry
combustion), hydrogen ion activity (with a glass electrode using a saturated
soil paste), and cation exchange capacity. Exchangeable potassium, calcium, and magnesium were displaced with ammonium acetate solution and
estimated by atomic absorption spectrophotometry (Black, 1965). Soil
moisture content was determined gravimetrically on samples brought to
equilibrium on a semi-permeable ceramic plate maintained at a pressure
differential of 1/3 and 15 bars.
Measurements of density, height, and grain yield of corn plants were
taken within three, four, and five 4-m transects randomly placed along rows
within the canopy, shade, root, and non-influence zones in mid-October
after the maize had fully matured. Grain yield was estimated by determining
the length and basal diameter of each ear along the transects using a ruler
and calipers. A grain yield per ear index was obtained by multiplying length
by diameter o f ears. This method was extensively used during the Puebla
Project to estimate yields of target maize fields (CIMMYT, 1974).
Three maize fields located 1.5-2 km east of Cuapiaxtla were chosen
to study the potential effects that typical maguey (Agave sp.) border plantings have on adjacent maize crops. The fields were 150 m x 50 m and bordered on both sides by a row of 1.5-2 m high maguey plants running the
length of the field. The outer maize rows were planted within a distance of
1.5-2 m from the border rows of maguey. After the maize had matured,
plant density and yield were measured along five random 4-m transects in
the outer two rows on both sides and center two rows of each field. In addi-

Corn Production in Tlaxcala

195

tion, growth and yield measurements were taken in various other tree-crop
associations, including corn fields cleared within both a sabino and ailite
(Alnusfirrnifotia) woodland and corn interplanted within apple and pearapple orchards.
The Influence o f Vegetational Diversity
on Insect Populations

Soil dwelling and foliage insect communities were monitored during

1982 and 1983 (from May-August) in five corn fields of varying degrees of
vegetational diversity: corn monoculture, corn strip-cropped with alfalfa, corn
mixed with fava beans, corn intercropped with apple trees, and corn growing in an agroforestry system composed of sabino, alite, and capulin trees.
On a weekly basis, 20 corn plants were randomly selected in the center
and border rows of each cornfield, and all arthropod species present on these
plants and the level of pest damage were recorded. Soil arthropod populations were monitored by randomly placing five pitfall traps in the center of
all fields. In the cornfields, adjacent to woodlands and alfalfa, five additional pitfall traps were randomly placed along the borders. The pitfalls were
filled with a mixture of 250 ml water and 100 ml antifreeze, and replaced
weekly. Traps were taken into the laboratory where arthropods were sorted
and counted (Baars, 1979).
RESULTS AND DISCUSSION
The Corn Agroecosystems

The agricultural systems of Tlaxcala are closely related to the "Tequexquinahuac" system, which is based on the production of a variety of crops
and animals assembled in several subsystems: (1) rainfed annual cropping
systems (parcela de temporal), (2) kitchen garden (huerto familiar), (3)
"agostadero," which includes fallow fields, secondary forests, and other areas
for gathering, (4) forest plots in gullies and hillsides for firewood and charcoal production, (5) confined management of pigs, poultry, rabbits, etc.,
and (6) free-grazing management of horses, mules, cows, oxen, and goats.
Our study focues on the dynamics of corn pi'oduction in "parcelas de temporal," where corn is grown associated with a number of annual, semiperennial, and perennial plants in polycultural and/or agroforestry patterns
(Fig. 2).
A common component of the borders of cornfields on hillsides is the
maguey, a multipurpose plant of the central valley, sacred to the early Indi-

196

Altieri and Trujillo

Fig. 2. A multiple-useagroecosystem,comprisedby capulin (Prunuscapulh), tejocote(Crataegus mexieana), pear-apple (Pyruscommunis), and maguey (Agave spp.).

ans (Benitez, 1953). About 130,000 ha of agricultural land in the state have
maguey planted in contour on sloping land or as part of a terracing system,
a practice that can substantially limit soil erosion (Cruz, 1949; Donkin, 1979).
Maguey is most valued for its role in the production ofpulque, a traditional
drink of the region.
Farmers also use maguey leaves as a rich organic amendment for the
soil, to make a strong fiber for making ropes, lassos, and collars for mule
teams, as a hot-burning, low-smoke fire source, and as forage for pigs. The
strong spines have been used as needles, and the outer skin of the leaves can
be peeled off and used in cooking. Magueys are also a source of the much
coveted "gusano de maguey" (Hypopta agavis and Aegiale hesperiaris), worms
considered to be a delicacy by many Mexicans; in 1983, the value of one kg
of worms matched the price of 10 kg of beef.
Trees are also conspicuous components of the agricultural landscape.
A number of different species, i.e., capulin (Prunus sp.), sabino (Juniperus
spp.), tejocote (Crataegus mexicana), peach (Prunus persica), and tepozan
(Buddleja americana), are integrated with corn in a variety of patterns resembling a veritable crop savannah (Fig. 3). These trees are not commonly planted
by farmers but are maintained along field borders or left scattered within

Corn Production in Tlaxcala

Fig. 3.

197

Corn planted within a woodland of sab!no (Juniperus deppeana) and capulin (Prunus
capuh), conforming a typical agroforestry pattern.

fields with corn planted right up to their base. Still higher tree densities can
he found where narrow fields have been cleared and planted to crops within
dense sabino and ailite woodlands. Trees and corn are also mixed in orchards,
and in fruit-growing areas such as Cuaxomulco, rows of trees including
apple, peach, plum, pear-apple, apricot, and walnut are c o m m o n l y interplanted with corn and alfalfa.
Our interviews revealed that farmers appreciate the multiple uses of
trees. They value the year-round vegetative co~er provided by trees for erosion control, the addition or organic matter (leaf litter) from semi-deciduous
trees, and the fruit of noncrop trees such as capulin, tejocote, and peach.
Roasted seeds of capulin are a delicacy and are sold on m a n y street corners
in towns and cities for additional income. Trees are also a source of fuel
wood and protective shade during the summer. In the case of orchards, farmers feel that by intercropping they utilize limited land and resources more
efficiently. Wooded areas are also valuable sources of livestock food, especially the lower strata of grasses and herbs growing under trees bordering
cornfields.

198

Altieri and Trujillo

Traditional Corn Management Practices

Cropping Patterns
Our survey revealed that farmers choose their cropping patterns based
on land resources available, the productive capacity of the land, traditional
customs, cash constraints, family labor, and their perception of overall risks
arising from natural and economic circumstances. In most areas, the major
cause of crop failure is unpredictable rainfall; thus, farmers stagger planting a n d / o r diversify cropping systems to reduce the effects of rainfall unreliability.
Corn intercropping systems are not commonly found in Tlaxcala as in
other areas of Mexico. The traditional corn-bean-squash polyculture is confined to selected areas inhabited by indegenous populations, particularly in
the Belen area and the drained-field zone of southwestern Tlaxcala where
soil moisture is not so limiting (Wilken, 1969). More common, however, are
mixtures of corn and frijol (Phaseolus spp.) or corn and fava beans (Vicia
fava). The planting of corn with frijol is done in the traditional way of sowing both corn and bean seeds in the same hole; however, planting designs
vary depending on the bean variety used. Usually, bush beans (Frijol ayocote) are planted at lower densities (about 8000 plants/ha), than most climbing
bean varieties (mateado, venturero, guia larga), which are usually planted
at very high densities (10-100,000 plants/ha). Fava beans are normally planted
between mounds of corn within a given row. Both types of beans are important food sources and are also recognized for their soil-improving qualities
("buen rastrojo"). The corn-fava bean combination provides the additional
benefit of risk aversion in areas to frost damage since fava beans are fairly
tolerant of frost conditions.
The rotation of corn and alfalfa is very c o m m o n in areas with available irrigation or with deep soils characterized by sufficiently high moistureholding capacity. Alfalfa is grown as animal fodder and planted with corn
in a strip-cropping arrangement, creating a mosaic pattern in the landscape
that changes every year, as corn is grown where alfalfa grew the previous
year and vice versa (see Fig. 2). Based on reliable agronomic data from similar areas, we estimated that, under Tlaxcala conditions, alfalfa could fix up
to 150 k g / N ha, almost sufficient to meet most of the N requirement of a
preceding corn crop (Sprague and Triplett, 1986). Other crops often grown
in rotation with corn are wheat and barley, which are valued for their suppressive effect on certain weed species. There are numerous variations in rotation patterns, including corn-barley (or wheat)-corn, corn-fava bean (or
frijol)-corn and corn-fava bean-barley-corn.

Corn Production in Tlaxcala

199

Soil Preparation

Farmers usually prepare the soil for planting around December. In sloping lands or very sandy soils susceptible to erosion, the soil is not worked
after harvesting but rather just before planting. Thus, crop residue from the
previous crop is left standing to protect the soil against early spring winds
and rain.
Soil preparation generally involves breaking up the larger soil clods,
the barbecho, and the mixing and leveling of the soil, the rastreo. There are
variations in the practice depending on soil type and availability of equipment, but the end result of loosening and leveling the upper 20-30 cm of
soil is the same. The equipment used includes discs, ploughs, harrows, or
a combination of these. If a team of animals is used, a plough is generally
used to loosen the soil while a railroad tie is often dragged for leveling.
Planting

The most commonly planted corn variety is the criollo blanco (white
seeds), which matures in about 6-7 months and gives the highest yield depending on environmental conditions. Many farmers, however, prefer to mix
several corn varieties in their fields, especially several types of early-maturing
varieties (violentos) including criollo amarillo and toluqueno (yellow seeds),
criollo negro, morado, or azul (blue-purple seeds), and another white-seeded
variety, monte alto de ocho carreras. These mature within about 4-5 months
after planting. The high genetic diversity resulting from such mixtures apparently provides some degree of risk aversion from diseases, pests, frosts,
etc. Farmers normally obtain their seeds by selecting the best grains (xinastli) from the previous year's harvest, but when the harvest is poor, farmers
buy seeds in the market or borrow from relatives and friends. A few farmers
have recently switched to hybrid seeds, but most recognize the low ecological amplitude and the high input requirements of these varieties.
Considering the unpredictable nature of the weather in Tlaxcala, the
decision of when to plant often represents a big gamble for the farmer. The
object is to begin early enough to take advantage of a long growing season
and to harvest before the early fall frosts. Thus, the average planting date
usually occurs in April, although many farmers choose to plant in March.
In particularly dry years, the first seeds may not be sown till June. Frost
and rainfall risks may be avoided by planting later with an early maturing
variety, although yields may be significantly lower. In years of late rains,
farmers often choose to plant barley or wheat rather than corn. In the case
of partial crop failure early in the season, most farmers will replant with

200

Altieri and Trujilio

the same type of seed or with early-maturing varieties in affected parts of

the field.
Because the appropriate planting time is so dependent on the weather,
farmers have developed unique and elaborate methods for predicting weather
conditions passed from generation to generation. The cabanuelas are one
of these predictive guides and apply to longer-term weather patterns. This
method consists of the interpretation of weather conditions during the initial or final 12 days of a given year, including fractions of certain of these
days. It is believed that the climatic conditions experienced on these days
or factions thereof represent the conditions to be expected during their corresponding months of the upcoming year.
Farmers have also developed effective ways to predict short-term
weather conditions. One of these is the observation of cow milk production, which according to farmers, changes with weather. The second involves
the observation of maguey plants. Sometimes referred to as "vacas verdes"
(green cows), they are said to produce more juice than normal with the
coming of rain. When the quantity of juice is less than normal and the interior of the bowl takes on a white color, dry days may be expected.
Planting is most often carried out manually with a shovel, although
more farmers are now using mechanical planting equipment. Furrows are
made with a plow and seeds generally planted at the bottom of the furrow
to a depth of 10-15 cm, depending on soil moisture. Many farmers affirm
that the depth the shovel penetrates depends on the amount of moisture in
the soil, suggesting that they use this method to indicate how deep to plant
at that particular time. Seeds are usually planted at a greater depth under
drier soil conditions, especially in sandy soils.
Spacing between plants within a given row is variable and depends on whether
seeds are planted in clumps (matas) or evenly spaced, on the availability of
water and nutrients, and, in some cases, whether the corn is grown for fodder or grain. Matas tend to be spaced about 80 cm apart with 2-4 seeds usually
planted together. Plants sown individually are normally 50-60 cm apart. Corn
planted for fodder is sown at much higher densities, with spacings of 25-30
cm within the row. Average distance between rows, depending on the size
of the plow, ranges between 50-90 cm.

Cultivation and Management o f Weeds

Three cultivation practices are carried out after initial planting: the escarda, the labra, and the segunda. The escarda is done when the young plants
have reached a height of about 10-15 cm and transfers soil from the adjacent
mound to the furrow, building up soil around the plants for support. The

Corn Production in Tlaxcala

201

furrows also collect water in the area of root extension. The labra usually
occurs when the plants have reached a height or about 25 cm. A slightly wider
and deeper plow is used to build additional soil around the plants. The segunda is the last pass, when corn is about 100-cm tall, which deepens the
furrow and builds up the mound higher and steeper to help guard against
lodging.
The segunda is also the last cultural measure taken to suppress the
growth of a variety of weeds that commonly invade cornfields, i.e., Simsia
foetida, Lopezia racemosa, Brassica campestris, Bidens spp., Cyperus spp.,
Chenopodium graveolens, Raphanus raphanistrum, and Lupinus elegans.
After this point, farmers recognize that these plants have little negative effect on corn growth (Moreno, Trujillo, and Ruiz, 1981).
Farmers do not refer to these plants as weeds (malezas) but rather as
"hierbas," "arvenses," or "quelites" (wild plants); therefore, many of these
plants are deliberately left in the fields as a second crop. Many of these quelites are important sources of minerals and vitamins such as calcium, thiamine, riboflavin, vitamin A, and vitamin C, thus improving the nutritional
quality of the diet (Bye, 1981). In fact, there is evidence that Tlaxcalan farmers
actually "sponser" wild plants in their fields through selective weeding
(Williams, 1985). Many species of Solanum, Jaltomata, and Physalis are so
adapted to the traditional patterns, perfectly mimicking the biological cycle
of crops, to the point that their fruits mature when crops are ready for harvesting (Fig. 4). In a Tlaxcala barley field, Williams (1985) estimated a yield
of 1325 kg of fruits from 4700 Solanum mozinianum plants/ha, with no apparent significant impact on barley production. In that area, fruit gathering
afforded a meaningful input to agricultural subsistence, since the fruits of
these "arvenses" are used for domestic consumption, restricted commercialization, and for ceremonial purposes (Wilkens, 1970). In the areas of Tlaxcala, Contla, and Apizaco, two species ofAmaranthus (A. Hypochondriacus,
and A. cruentus) are still grown associated or in rotation with corn, as farmers still recognize the ritual as well as nutritious (13-18% protein) importance of "alegria."
Chicle (Asclepias spp.) is traditionally used for making chewing gum
and as a hemorrhoidal cure and can be commonly seen "sponsored" within
many fields. Nopal (Opuntia spp.) is probably the most widely used of the
"wild" plants, although it is planted as a crop in some areas. The "leaves"
are peeled and cooked as a vegetable and the fruits (tuna) are eaten raw.
They are also used as soil stabilizers and have been planted extensively on
terrace risers east of Tlaxcala City. In certain areas such as Tequixquitla and
Altamira, a variety of grass species are used for roofing material in homes
and storage facilities. Table I lists several wild plants and their corresponding uses by TlaxcaIan farmers.

Altieri and Trujillo

202
i

corn

barley

o
zk

$. procumbena

/x

S. c h e n o p o d l f o l l a

~
I

[]

0
!

S.mozlnlanum

0
I

[]

[]

e0.

<

60, ,.osT

so.~
4o.I

9 0
1

0
I

CC 30
20

9
!

~,....~

"

;
"

t..

"

'"

9-

:
:
:

8 6

-:

":

~,:

70-

[] 0

.-Jj

90.

,
9

[] 0

..A...... . . . . .

1o0.

1110.20'

n:=

""

,.osT

4.
".

".

=
L,o
/

15

,,:,

10 I-"
,

IJ ] F I M ] A I M I J I J

IAIS

]01NIDI

Fig. 4. Agricultural and phenological cycle of corn, barley, and wild Solanaceae (Solanum
spp.) in Tlaxco, Tlaxcala (Williams, 1985). (1 = sowing, 2-4 = cultivation practices, 5 =
corn piling, 6 = harvest, 7 = barbecho (fallow), 8 = mulching,/x = spontaneous emergence, O = flowering, [] = fruiting, and 9 = fruit maturation).

Farmers derive other benefits from the presence of tolerable levels of

wild plants in their fields. For example, frijolillo (Lupinus spp.) and colza
(Brassica campestris) are at times intentionally left within corn fields, playing a beneficial role on the dynamics of the corn pest frailecillo (Macrodactylus sp). Our observations rvealed that frailecillo consistently fed more often
on the flowers of colza and lupine than on corn flowers, indicating the potential importance of these plants as trap crops. Many plants are also collected
after crop harvest, to be worked into the soil later during postseason cultivation, to enhance organic matter content in the fields.

Fertilization
The use of chemical fertilizers has increased over the last 15 years due
to available credit. C o m m o n fertilizers used include superphosphate of cap

Corn Production in Tlaxcala

203

Table I. Non-Cultivated (Wild) Plants Utilized by Farmers in Tlaxcala, Mexico

Scientific name

Common name

Uses

Amaranthus hypopondriacus ~
Amaranthus hybridus
Argemone mexicana
Aristida divaricata

Ingredient for making candies

Forage
Remedy for eye and skin infections
medicinal

Artemisa absinthius
Bidens pilosa
Brassica campestris
Cedronella mexicana
Chenopodium ambrosioides

Alegria
Quelite
Chicalote
Hierba (Pasto)
de la virgen
Ajenjo
Rosilla
Colza
Toronjil
Epazote

Chenopodium graveolans
Cyperus sp.

Epazote de zorillo
Tule and Coquillo

Encelia mexicana
Hedeoma piperita

Acahual
Yerbabuena

Heterotheca inuloides

Arnica

Ipomoea starts

Cacaxtlapa

Marrubium vulgare

Marrubia

Matricaria parthenium
Opuntia spp.
Origanum vulgare
Ruta graveolens

Manzanilla
Nopal
Oregano
Ruda

Solanum mozinianum

Tlanochtle

Thymus vulgaris

Tomillo

Digestive; kills intestinal parasites

Forage
Bird food, industrial oil source
Appetizer, digestive
Flavoring, remedy for intestinal
parasites
Intestinal parasites, diuretic
Weaving of "petates" (bed mats)
and baskets
Forage
Intestinal disorders and bronchitis,
antispasmodic
Helps blood circulation, recovers
muscles from contusive damages
Relieves stomach and menstrual
pains, sedant
Appetizer, remedy for diarrhea,
bronchitis, and live inflamations
Appetizer, antispasmodic
Edible fruit and leaves
Condiment, antiseptic
For hepatic and intestinal pains;
used to control skin lice
Berries used for human consumption and ceremonial purposes
Cures gastroenteritis, antiseptic

"This plant is cultivated in some areas, although plantings rarely exceed 0.1 ha.

c i u m o r a m m o n i u m , u r e a , a n d a n 18-46-0 m i x t u r e . I f a f f o r d a b l e , f a r m e r s
a p p l y fertilizer twice d u r i n g the season. T h e first a p p l i c a t i o n c a n be either
at the t i m e o f p l a n t i n g o r d u r i n g the escarda. T h e s e c o n d a p p l i c a t i o n is c o m m o n l y d o n e d u r i n g the segunda. T h e fertilizer is d i s t r i b u t e d b y l a n d , with
a given p o r t i o n d r o p p e d either at the base o f each p l a n t o r g r o u p o f p l a n t s
( m a t e a d o ) , o r in strips a l o n g the f u r r o w (en b a n d a ) . M a n y f a r m e r s feel t h a t
c h e m i c a l fertilizers are easy to use, a n d a r e effective even with little rainfall.
H o w e v e r , o t h e r s c o m p l a i n t h a t fertilizers are costly a n d last for o n l y 1 year,
m u s t be a p p l i e d each year, a n d , in the long t e r m , i m p o v e r i s h a n d w e a k e n
the soil because o f the lack o f organic m a t t e r i m p o r t . In a d d i t i o n , m a n y f a r m ers indicate t h a t excessive use o f fertilizers " b u r n " the soil a n d create p r o b l e m s
with salinization; h o w e v e r , we d i d n o t o b t a i n scientific evidence to s u p p o r t
these claims. B a s e d o n o f f i c i a l d a t a , it seems t h a t f a r m e r s were b e t t e r o f f

204

Altieri and Trujilio

Table II. Production Costs and Yields of Corn under Different Technological
LeveJs in Tlaxcala, Mexicoa
Cost
Yield
Net gain
(pesos/ha) (kg/ha)
(pesos/ha)
Improved seed, with fertilizer
Improved seed, without fertilizer
Criollo seed, with fertilizer
Criollo seed, without fertilizer

3495
2074
3130
2045

2534
1537
1932
1284

960
990
720
740

aUnder rainfed ("temporal")conditions, averageof 3-yeardata 1973-1976. Banrural (1976), Costos de producci6n de once cultivos bfisicos, Fideicomiso para
la realizaci6n de Estudios de Desarrollo Agropecuario.

economically by not using synthetic fertilizers regardless of the type of corn

variety used. As Table II suggests, although, in the short term, fertilizers
increase yields, their use does not compensate for the higher costs of production that these inputs imply.
Low income farmers who cannot purchase commercial fertilizers sustain soil fertility by collecting nutrient materials f r o m outside the farm, such
as manure collected f r o m pastures or enclosures in which animals are kept
at night. This organic material is supplemented with leaves and other plant
materials collected from nearby forests and fallow fields. Other farmers recycle nutrients a m o n g the various enterprises on the single farm, composting
waste plant materials with household wastes and manure from livestock. Older
leaves f r o m the maguey plant are especially favored, and, in the primary
pulque-producing areas, large quantities of leaves are collected in piles with
other debris and soil and allowed to decompose before being used as a rich
amendment. As the manure effectively provides nutrients for about 3-4 years,
farmers have developed yearly rotations, applying available organic material to low-fertility areas of their fields. A familiar sight at the beginning of
the growing season is mounds of manure evenly distributed over fields before being spread and incorporated into the soil (Fig. 5).
One method used to take advantage of soil enrichment by decomposing plant remains is the practice of rotating the placement of maguey borders. Once the maguey stop producing and begin dying back, they are removed
and a new row planted where the center of the field used to be. The following year, corn is planted over the area of the previous maguey border, thus
taking advantage of the organic matter accumulated over the years.
Another very old method of soil enrichment is the use of catchment
ditches (cajetes) among field borders at the bottom of the slopes. These ditches
collect runoff and debris, which is periodically returned to the field to replace
lost nutrients and organic material.
The most valuable characteristics of organic manure recognized by farmers are its role in building up the soil, retaining soil moisture, and the fact

Corn Production in Tlaxcala

205

Fig. 5. Piles of manure (abono) ready to be spread over a field before corn planting in late
March.
that its effectiveness lasts for about 3-5 years, all critical properties in sandy
soils of low organic matter and low CEC. Farmers also indicate some
problems with the use of manure, such as the production of "hierbas" the
first year due to the number of weed seeds brought in with it and the favoring of gallina ciega (Phyllophaga spp.), a saprophagus larval pest that is harbored in organic waste.
Another strategy used by farmers to sustain soil fertility is to exploit the ability of the cropping system to reuse its own stored nutrients. Based on
current information on nutrient dynamics in agroforestry systems (Nair, 1984),
it is reasonable to expect that minerals lost by annual crops in Tlaxcala are
rapidly taken up by perennial crop plants. In addition, the nutrient-robbing
propensity of corn can apparently be counteracted by the enriching addition
of organic matter to the soil by associated trees. Soil nitrogen is increased
by incorporating legumes (fava beans or alfalfa) in the corn-cropping system.

Pest Control
The most important insect pest of corn is the scarab beetle Macrodactylus sp. or fraileciUo, which feeds on newly formed female flowers, thus
impeding grain formation. Armyworms (Spodoptera spp.), cutworms (Prode-

206

Altieri and Trujillo

nia sp., Agrotis spp., and Fenia spp.), aphids (Ropalosiphum spp.), and gallina ciega sporadically cause damage.
Farmers believe that insect pest outbreaks are instrinsically related to
climatic conditions. This is particularly true for frailecillo, armyworms, and
red spider mites which tend to be more abundant during drier periods. Farmers also feel that the excessive use of chemical fertilizers and not preparing
the soil in time leads to increased numbers of pests. Some farmers associate
the phase of the moon at the time of planting with the occurrence of pests,
while still others believe that the presence of pests is a consequence of their
bad personal behavior, accepting them as a divine punishment.
Some explanations given decreases in pest numbers include climatic conditions such as low temperatures and strong and/or continuous rain, increasing plant maturity, the coming of a full moon, and certain religious days.
Despite credit support and technical assistance from agricultural extension, pesticide use in corn is still minimal, mostly limited to control severe outbreaks of frailecillo, armyworms, and rodents. Tlaxcalan farmers
practice a number of cultural control methods, such as the manual collection of frailecillo from plants, and the direct application of lime onto frailecilIo in the morning when dew is still present. Most farmers cultivate after the
harvest or before planting to effectively destroy soil-borne insects. Other farmers use measures based on religious beliefs. For example, they collect a number of frailecillo and take them to the local where they are exorcized by one
of the priests and later released back into the fields where they were collected.
In the valley areas, farmers have problems with mice and rats, whereas
in the slopes of hill, they tend to have more problems with gophers and squirrels. Mechanical traps are often used to reduce rodent populations. Farmers
also state that the planting of ayacote (Phaseolus coecineus) within fields
is effective in preventing the establishment of gophers, apparently because
these plants exude toxic substances from their roots. Many farmers recognize the importance of birds and snakes in the control of rodents. However,
they do not seem to recognize the importance of insect predators and parasites in the maintenance of arthropod balance in their fields.

Harvesting
The most common harvesting method is the "amogotado," which involves cutting the corn plant at the base with a machete after the ears have
matured. Plants are then collected and piled together in the form of small
teepees (mogotes) and left to dry for 1-2 months, after which time the ears
are picked and put in large burlap bags and the plant remains collected for
forage. Some farmers, however, let the corn ears dry on the plant in the field

Corn Production in Tlaxcala

207

and later pick them directly. This method is referred to as "dobla": the plant
is broken toward the middle, doubled over, and left to dry.
Most farmers prefer the amogotado method, because by cutting and
stacking the corn before it has dried completely, they are able to plow their
fields sooner than they could otherwise. This enables them to incorporate
the remaining noncrop plants into the soil when they are still green. Cutting
the plants early also conserves the leaves, which otherwise might fall off when
harvesting if the plants were dry. However, corn stacked in piles is most susceptible to disease after late rains as well as to damage by mice and rats.
Corn grain is usually stored still attached to the cob in specially built
storage facilities (cuexcomates) or in one of the rooms of the house. To protect the grain against potential pest damage, farmers find that temperature
and humidity are the two critical factors to control. In the cooler regions
of higher elevations there are few problems with storage pests, but in warmer areas, some farmers must resort to chemical insecticides, e.g., Malathion,
Lindane, etc., for control of weevils (Circulionidae) and Pyralid moths.

E c o l o g i c a l Interactions

The Effects of Individual Trees on Soil

and Microclimatic Properties
In the soil study plots, a decreasing gradient in all soil properties measured was observed with increasing distance from the trees (Table III). Soil
nitrogen, phosphorus, CEC, and calcium gradients were especially
pronounced in the corn-capulin systems, suggesting that capulin had a greater
degree of influence on soil properties than sabino. Available phosphorus increased 4- to 7-fold and nitrogen 1.5- to 3-fold under the trees (Fig. 6). Total
carbon, calcium, magnesium fold, and CEC increased. Potassium levels
decreased with distance from the trees, although its concentration was higher
under sabino canopies, soil pH and percent moisture were also slightly greater
under tree canopies. Both trees appeared to influence surface soil properties
within a radius of 6-10 m from their base (Farrell, 1984).
The redistribution of nutrients with litterfall and the resulting accumulation of organic matter under the trees are probably the only significant factors determining the observed trends of soil chemical properties (Zinke, 1962).
The higher concentration of carbon beneath capulin and sabino followed
an expected trend due to the accumulation of litter by the trees, and a general lack of organic matter used in the management of these agricultural soils.
The distribution of nitrogen under the trees also seemed correlated with car-

7.4
6.9
6.5
6.6
6.7
6.4

Sabino
2
6
10
14
18
22

~See Farrell (1984).

bMean values _+ S.E.

+_

6.6
6.4
6.2
6.0
5,8
5.9

Ca pu lin
2
6
10
14
18
22

.26 b
.28
.18
.00
.05
.09

+__ .01
.15
.13
-+ .05
.38
.21

pH

Distance(m)

-+
_+

.63 -+
.51'
.36
.37 _+
.40
,40

1.34
.73
,60
,47
.43
.50

.10
.07
.02
.06
.01
.02

.04
.23
.20
.12
.11
.12

TotalC

.044
.038
.030
.033
.031
.030

.091
.051
.041
,031
.028
.033

+
_+

+
+_

.008
.001
.000
.001
.001
.001

.001
.012
.009
.005
.003
.005

Total N

2.5
1.3
.6
1.4
2.2
1.5

17.3
6.6
3.4
2.3
2.0
2.2

1.06
.18
.04
.39
.83
.53

.58
2.19

.97
1.01
1.00

.87

A va i l a bl e P

6.4
6.0
5.5
5.7
5.5
5.6

7.0
5.2
4.8
4.1
4.1
4.6

+_
+

.69
.37
.69
.03
.08
.27

.48
+ 1.26
_ 1.00
_+ .85

.71
+ .60

CEC

6.6
5.1
4.1
4.2
4.5
4.3

6.5
3.9
3.0
2,3
2.1
2.6

_+

Ca

.10
.49
.17
.50
.36
.29

.05
.69
.35
.20
.32
.64

Exchangeable

1.24
.97
.85
.97
.98
.89

1.29
.69
.52
.40
.34
.34

.23
.08
.11
.09
.09
.08

.15
.01
.04
.14
.03
.03

Mg

.84
.58
.48
.47
.56
.48

.58
.26
.19
.17
.15
.18

.13
.22
.00
.01
.03
.02

.04
.00
.01
.02
.01
.03

9.2
8.4
8.2
8.2
8.4
8.3

11.1
10.4
9.6
8.8
8.1
8.8

_+
_+
+
-+

.56
.22
.14
.08
.39
.18

1.0
1.3
.75
.14
.71
.62

(1/3 bar)

%H20

%H20

5.4
5.1
5.0
5.2
5,1
5.1

+
+
_+

.52
.25
.28
.01
.01
.12

.22
.21
.11
_+_ .12
.12
.00

(15 bars)
5.9
4.4
3.9
3.6
3.4
3.7

Table I I I. Surface Soil Properties in a Tlaxcalan C ornfi e l d at Six Distances from the Base of Individual C a p u l i n a nd Sabino Trees ~

Corn

Production

in Tlaxcala

209

0.10
CAPUUN

0.09

0.08
0.07
0.06
Z
w

0,05

0
Z

SABINO
0.04
0.05
0.02
0.01
0.00
0

10

14

18

22

24

DISTANCE FROM TREE (meters)

20
19
18
17

CAPUUN

16
15
14

m
~
o
i

11
lO
9

~0

1I1

7
6
5
4.
3
2
1

1
2

L
6

I
I0

14

18

22

24-

DISTANCE FROM TREE (meters)

Fig. 6. C h a n g e in m a j o r surface soil nutrients in a corn-woodland agroforestry system, with increasing distance from individual capulin (Prunus capuh) and sabino (Juniperusspp.) trees: (a) nitrogen,
and (b) phosphorus OVarrell, 1984).

210

Altieri and

Trujillo

bon accumulation. Accumulation of exchangeable cations also appear to be

closely associated with organic matter.
Phosporus markedly accumulated beneath the capulin trees, corresponding with phosphorus amounts deposited with leaf litter. In addition, the higher
pH found beneath the trees compared to in the open seemed to account for
the increase in phosphorus availability in that area.
Soil moisture differences at the 0- to 15-cm depth were most pronounced
around capulin where moisture was greatest under the tree canopy, somewhat less in the shade-west zone, and lowest in the remaining zones. No differences were observed between the shade-east, root, and non-influence zones.
Around sabino trees, surface soil was wettest under the canopy, slightly
drier in the two-shade and root zones and driest in the non-influence zone.
In all zones except under the capulin canopy, soil at 15-30 cm was wetter
than surface soils. No differences were observed between any of the zones
at this depth, indicating that tree influences diminish with depth.
The increased capacity of the soils under capulin and sabino to retain
water compared to soil in the open is due, in part, to the higher levels of
organic matter beneath the trees. More important to the understory corn is
the amount of water available for use (measurable difference in moisture
between 1/3 and 15 bars), which was somewhat higher beneath the trees than
away from them.

Corn Growth and Yield in Agroforestry systems

Corn yields varied considerably among the sampled agroforestry systems (Table IV); however, since the data proceed from nonreplicated farmers' fields, the effects of trees on corn yields cannot be statistically separated
from other sources of variation, such as soil, microclimate, and management
differences between agrofoestry systems, which can also determine significant effects on yields. Nevertheless, the data provide a tentative description
of the yield potentials of each system. No significant differences (p = .05)
were found in either plant height, density, or grain yield between corn growing along the edge (yield index = 729 _+ 102) or in the center (734 +_ 279)
of our study fields bordered by rows of maguey. The characteristically wide
border strips along which the maguey are planted (usually 4 m) ensure a 1.5to 2-m distance between the last row of corn and the maguey border. Thus,
although maguey plants have rather extensive superficial roots, they do not
reach far enough into the field to have a substantial effect on corn production within the bordered fields.
Development and yields o f early (March) and late (May) planted corn
were measured in three zones in two sabino woodland fields. Zone 1 was
located along the field borders, while zones 2 and 3 were progressively far-

Corn Production in Tlaxcala

211

Table IV. Average Yield, Height, and Density of Maize Grow-

ing in Association with Various Perennial Plants in Agroforestry Systems

Height Density(number
System
Yiel& (cm)
of plants/ha)
Apple orchard
Pear-apple orchard
Sabino woodland
Early-planted
Late-planted
Ailite woodland
Maguey
NI (capulin)b
NI (sabino)

597
912

201
245

34,515
33,638

1127
573
617
734
1005
1073

212
173
247
162
192
191

51,772
40,950
32,175
45,922
57,623
40,950

"Yield expressed as a relative index per 4-m transect (CIMMYT,

1974).
bYields from the non-influence(NI) zonearound capulin and sabino are included for comparison and are regarded as equivalent
to monoculture yields.
ther from the trees into the center of the fields. In the early planted field,
the rate of maize growth was staggered between the zones with maize in zone
3 maturing first, followed by zone 2 and zone 1. There were no differences
in final grain yield between zones 2 and 3, while maize in zone 1 produced
about 33% less (750 _+ 95.1) and center (573 + 90.9) zones in the late-planted
fields. Throughout the season, corn growth was slowest along the eastern
field borders. Corn along the western borders matured sooner than in the
other sectors of the field. Grain yield was lower in the eastern border than
in the other two zones. No differences in yield were found among corn plants
grown next to a border of ailite trees (491 +_ 88.3) and corn plants grown
in the center rows (554 + 116) of the same field. However, there were significant differences in plant height; maize along the center row away from
the trees was 12% shorter than maize near the trees (237 cm).
Yield and height of corn plants growing near apple trees (516 _+ 59.3)
or 5 m away in the center (597 + 191) o f the interrow strips were similar.
Yields of corn intercropped within pear-apple/peach orchards compared
favorably to yields in other sampled agroforestry systems (Table IV).

Arthropod Dynamics in the Vario~ts Cropping Systems

Foliage Arthropods. Depending on the degree of vegetational diversity
and the management intensity of the fields, populations of beneficial insects
varied in species numbers and in relative abundance. Dominant predators
included lady beetles (Hippodamia convergens, H. Koebeli, Coccinella
nugatoria, and Scymnus sp.), malachiid beetle (Collops sp.), several species

212

A l t i e r i a n d Trujillo
30
28
26

24
22
20
18

16
14"
12

1086420

,~ ~ u
9

10

11

12

13

14

15

Weeks

after planting

70

60

50

;~

40

3o

20

10

10

11

12

13

14

15

Weeks

offer

planting

Fig. 7. Cumulative mean densities per 200 corn plants of: (a) the pestiferous scarab beetle Macrodactylus sp., and (b) the lady beetles Hippodamia convergens, and Coccinella nugatoria, in
a range of corn-cropping systems. ([] = corn monoculture, + = corn-lava bean intercropping, Lk = corn-alfalfa strip cropping, ~ , = agroforestry system of corn under a tree layer
of sabino and ailite trees, and = corn intercropped within an apple orchard).

Corn Production in Tlaxcala

213

of Carabidae, Staphylinidae, predaceous Hemiptera such as Orious sp.,

Nabis sp., predaceous Diptera of the families Syrphidae and Dolichopodidae, and the common earwig (Doru sp.). Spiders of the families Lycosidae,
Argiopidae, Tethragnatidae, Salticidae, and Thomisidae were also commonly
found on the plants.
The various vegetational arrangements under which corn was found
affected the arthropod fauna in various ways. Some systems exhibited particularly high populations of corn-feeding insects. For example, in our sampled fields, the scarab pest, Macrodactylus sp., was predominantly more
abundant in the corn-alfalfa strip-cropping system, whereas in the corn system growing adjacent to ailite woodlands it reached the lowest densities (Fig.

7).
The only system that exhibited high densities of several entomophagous insects was the corn-fava bean-squash polyculture. Populations of coccinellid beetles and of a crab spider species (Thomisidae) were higher in this
polyculture system than in corresponding monocultures (Fig. 7b). The predator Orious sp. (Hemiptera: Anthocoridae) was noticeably more abundant
in the corn-orchard and corn-alfalfa systems than in any other system.
The proximity of alfalfa strips to corn had a fundamental effect on
the occurrence of all arthropods. As observed in Fig. 8, coccinellids

18O
170
160
150
140

nter,

I30
120
110

109

Weeks After Planting

90
80
70
60
50
4-0
50
20
10
0
A

Fig. 8. Seasonal abundances of (a) predaceous: Orius, (b) CoccineUidae, (c), Collops malachiid
beetles, and (d) the pest, Macrodactylus on corn plants located in rows immediately adjacent
to alfalfa strips and in rows located in the center of the field away from the alfalfa. The line
graph in the upper right corner illustrates the seasonal abundance patterns of Coccinellidae
in the border and center rows of corn.

214

Altieri and Trujillo

(Collops and Orius) were seasonally more abundant (about 30~ on corn
plants located in the row adjacent to alfalfa than in the center of the field,
away from the alfalfa strip. Macrodactylus, however, showed the opposite
trend. In systems where alfalfa was periodically cut, we did not observe these
abundance gradients. Apparently, cutting forces arthropods to disperse, and
through redistribution they attain similar densities throughout the cornfield.
Soil-Dwelling Arthropods. Pitfall catches yielded significantly higher
numbers of lycosid spiders in the corn-alfalfa system than in any of the other
corn systems (Fig. 9a). Spider catches remained at similar low levels in all
other systems, although at particular times spiders were caught more frequently in the corn-orchard system and in monocultures. Similarly, as in the case of
foliage predators, substantially more lycosid spiders were caught in pitfalls
placed in corn rows adjacent to alfalfa than in the center rows of the cornfield (Fig. 9b). Ground beetles reached highest densities in the corn-fava beansquash polyculture, apparently encouraged by the shelter provided by the
squash leaves.

DISCUSSION AND CONCLUSIONS

Many of the agricultural systems and practices currently used by Tlaxcalan farmers have evolved over many. years of adapting to a fragile environment with soil, climatic, and biological limitations to food production
(Williams, 1985). Other practices, however, have been more recently adopted as a result of increasing exposure to modern agricultural technologies.
Thus small farmers in Tlaxcala are in a state of transition from the traditional to the modern. Our surveys indicate that the rationales behind traditional farming practices are often rooted in a sophisticated understanding
of local environmental conditions and are closely associated with the persistence of a strong cultural tradition. Given the available resources and the
production constraints, these systems and practices often represent extremely appropriate forms of land use (Toledo, Carablas, mapes, and Toledo,
1985).
Although the described systems may bear close resemblance to systems
employed by other peasants in Mexico (Gliessman et al., 198 I; Wilken, 1970),
they differ from some traditional models in a number of ways. Notably, corn
is grown under a number of temporal (rotations) and/or spatial arrangements
(intercropping, strip-cropping, and agroforestry) instead of in a cyclic polyculture or agroforestry pattern characteristically used by farmers in the
lowland tropics (Beets, 1982). Tlaxcalan systems are genetically diverse, and
this diversity results from interactions far more complex than the random
planting of numerous corn varieties. Genetic diversity is also enhanced in
the fields as a result of the protection and encourgement of "wild plants"

Corn Production in Tlaxeala

215

280
260
240
220
200
180
o

;E

160
140

120

1O0
80
60

20

+---

-'49~

~ ------7

Week

Fig. 9. (a) Numbers of lycosid spiders caught in pitfall traps placed in corn strip-cropped with

alfalfa (A), intercropped with fava bean (+), intercropped with apples (@), and grown in
monoculture ([]), and (b) numbers of lycosid spiders caught in pitfall traps placed in corn rows
bordering alfalfa strips and in rows located in the center of the field.

that, in addition to firewood and medicines, provide dietary diversity. Thus,

the systems serve as sources for local household needs and also provide cash
returns f r o m the sale of the various products.
The data presented above outline some of the interactions of human,
management, and biological factors contributing to the ecological dynamics
of corn-cropping systems in the area. Apparently, growing corn in a multiple array of vegetational designs is a strategy well adapted to the plethora
of microenvironments to which corn production is subjected and an
effective way to cope with seasonality and variability.
Although the farmers may not be aware of it, we found that the production of corn in polycultures and agroforestry patterns triggers a series of ecological interactions with important consequences for insect pest management
and soil fertility relations. We recognize, howe~er, that our data proceed from
what m a y be considered "a handful of maize fields," and therefore we do
not pretend to make broad generalizations about the ecological dynamics
of corn agroecosystems. This would require more refined and well-controlled
comparative experiments under a range of site-specific conditions. Our specific quantitative analyses were intended to illustrate some key biological interactions that result when Tlaxcalan farmers assemble their fields in various

216

Altieri and Trujillo

RAIN
~.

WIN D . - ~ - - - - ~ . ~

~ Radiantenergy
ASSOCIATED ; Air movement
MiCROCLIMATIC Evapotransldr
CONDITIONS I ,~MSX,alr temp.

SOLAR
RADIATION

I.TCA rNO::T,n

~..~

EFFECTSON ~',1.LOdp,ngo.
ASSOCIATED |
leewardside
Slower development
In m

NU'mlI~IT

It Cation exchangecapacity /" t [~/ ~ \ V

A~SOCIA'I--cD ~J*~Nutrient concentrations
i
-fp+l

1,Sol, .... t.r.

/ ~ Moisture retention
| ~Surfaca temperature

Soll atlblllzlltlon

Fig. 10. A conceptualmodel of the potential influenceof trees on the growing environment
of corn in agroforestry systems of Tlaxcala (Farrell, 1984).
patterns. Our data suggest that, at least in our study fields, trees integrated
into cropping systems modified the growing environment of associated understory corn plants, influencing their growth and yields. Figure 10 depicts
various ways in which we assume selected trees modify soil and microclimatic factors. The greatest modification apparently results from the interception of solar radiation by tree canopies and deposition of leaf litter, but
the above-ground tree structure also plays a significant role in intercepting,
collecting, and redistributing precipitation, reducing wind velocities, and contributing chemical elements to throughfall precipitation. Below-ground, extensive root systems help to stabilize the soil, draw nutrients to be deposited
later on the soil surface with litterfall, and contribute organic matter to the
soil as a result of root decomposition.
With decreasing distance from trees, we observed an increase in soil
organic matter, total nitrogen, available phosphorus, exchangeable potassium, calcium, and magnesium, cation exchange capacity, and water holding
capacity. Soil moisture was also greater beneath the tree canopies. The
reduced amount of solar radiation reaching the ground beneath the tree canopies resulted in lower evapotranspiration and a marked reduction in surface
soil temperatures compared to the open cornfield. The trees also acted as

Corn Production in Tlaxcala

217

a buffer against air temperature fluctuations and, as a result, lower maximum and higher minimum temperatures were found beneath the tree canopies than in the open (Farrell, 1984).
Certain tree species affected maize yields more than others. Such trees
as apple, pear-apple, and peach, with smaller crowns and lower stature, appeared to have a minimal influence on yields; thus, substantial harvests were
obtained from maize intercropped in fruit orchards. Germination was reduced
near capulin and sabino trees in certain circumstances, and a reduction in
yields was observed only directly beneach the canopies of these trees despite
the fact that the soil near these trees was nutrient-rich. We feel that these
differences may be due to sunlight a n d / o r water differentials, an aspect that
needs to be experimentally analyzed. Although corn monoculture systems
in the region may exhibit significantly greater grain yields per hectare than
most of the surveyed traditional agroforestry systems, when other considerations such as risk, multiple protective, and producive tree functions, and
returns from the most limiting input, e.g., labor and capital, are examined,
the yield superiority of the monoculture becomes less clear-cut and attractive.
Although these results describe tree influences on crop development for
a single season, long-term effects are not indicated. Thus, in further evaluating the influence of trees on associated crops, the long-range effects, especially the trees' contribution to soil enhancement through organic matter input
and their potential role in regulating faunal communities in these
agroecosystems, should be considered. Such data can provide important information for determining which trees to encourage in fields, in what spatial designs, and for what purposes.
Our data on arthropod populations suggest that densities of the pest
Macrodactylus and of foliage and soild predators fluctuated depending on
the arrangement of crops in time and space, the composition and abundance
of noncrop vegetation within and around fields, the surrounding environments, and the type and intensity of management.. Thus, Macrodactylus sp.,
predaceous beetles, lycosid spiders, and Orius sp. responded depending on
their degree of association with one or more of the vegetational components
in the systems. For example, the abundance of insect predators and spiders
in corn depended greatly on the presence and phenology of adjacent alfalfa
strips. A greater abundance of predators in corn rows adjacent to alfalfa
strips corresponded with a lower incidence of Macrodactylus in those rows.
Since alfalfa serves as a major source of natural enemies, correct designs
of corn-alfalfa strip-cropping systems and proper timing of alfalfa cuttings
could greatly increase the abundance and efficiency of beneficial arthropods
in corn systems. Adoption of such designs would be confined, however, to
areas with available irrigation, since alfalfa requires abundant water for
growth. In addition, infestations of this pest on corn can be ameliorated by

218

Altieri and Trujillo

encouraging other host plants within or around the fields (Brassica spp., and
Lupinus spp.) that serve as trap crosp, enticing Macrodactylus away from
corn, as traditionally done by local farmers. Also the corn-fava bean-squash
polyculture apparently exhibits inherent elements of crop protection as found
by Letourneau (1983) in her studies in similar systems in Tabasco. We are
currently conducting further research in Huamantla to determine optimal
planting dates, planting density, and spatial designs of the component crops
of this polyculture, for improved biological control of key pests (Altieri and
Letourneau, 1982).
Since corn in Tlaxcala is managed over a range of energy inputs, levels
of vegetational diversity and successional states, variations in arthropod dynamics are likely to occur; thus, their populations will be difficult to predict
over broad areas. However, studies such as these will help us identify spatial
and temporal conditions under which low pest potentials can be expected.
In summary, we feel that there are several stabilizing elements inherent
to Tlaxcalan agriculture that should be preserved:
1. The farmers' tendency to manage polycultures and multiple purpose
systems, both small- and relatively large-scale.
2. The rational utilization of hillsides and slopes through permanent
terracing systems, especially managing Agave spp. borders.
3. The maintenance of genetically diverse fields apparently not only
reduces the threat to crop loss, but also constitutes important in situ repositories of crop germplasm.
4. The myriad examples of resource management and appropriation
systems to deal with site-specific environmental constraints, e.g., unique
watering and drainage systems (Wilken, 1969), the "cajete system," and the
large-scale canals bordering maize fields mimicking large "chinampas" found
in the valleys near Nativitas (see Toledo et al., 1985). So far, these local
agricultural skills have been underutilized, but constitute potentially profitable
resources for agricultural development.

ACKNOWLEDGMENTS
We thank the members of the Direcci6n General de Sanidad Vegetal
(SARH), Huamantla for their generous assistance and cooperation during
this project. A special thanks to G. Larrogaiti, e. Hermindez, G. Rangel,
and M. Mufioz for their field assistance, to M. Massion for the drawing of
figures, and to Dorothy DeMars for typing the manuscript. We also thank
John G. Farrell, Agroecology Program, University of California, Santa Cruz,
for allowing us to use some of his thesis data on the role of trees in Tlaxcalan agroecosystems.

Corn Production in Tlaxcala

219
REFERENCES

Alcorn, J. B. (1981). Huastec noncrop resource management. Human Ecology 9: 395-417.

Alcorn, J. B. (1984). Huastec Mayan Ethnobotany. University of Texas Press, Austin, Texas.
Altieri, M. A. (1987). Agroecology: The SeientificBasis o f Alternative Agriculture. Westview
Press, Boulder, Colorado.
Altieri, M. A., and Anderson, M. K. (1986). An ecological basis for the development of alternative agricultural systems for small farmers in the Third World. American Journal o f
Alternative Agriculture 1: 30-38.
Alverson, H. (1984). The widsom of tradition in the development of dry-land farming: Botswana. Human Organization 43: 1-18.
Anonymous (1981). Sintesis Geogrdfiea de Tlaxcala. Secretaria de Programaci6n y Presupuesto,
Tlaxcala, M6xico.
Anonymous (1984). Resultados de la Producci6n Agricola: Tenencia de la Tierra 1982-1984.
Secretarla de Agricultura y Recursos Hidraulicos, Mexico.
Baars, M. A. (1979). Patterns of movement of radioactive carabid beetles. Oecologia 44: 125-140.
Beets, W. C. (1982). Multiple Cropping and Tropical Farming Systems. Westview Press, Boulder,
Colorado.
Benitez, F. (1953). Los Primeros Mexicanos. Ediciones Era, Mexico.
Black, C. A. (ed.) (1965). Methods o f Soil Analysis. Part 2. Chemical and Microbiological Properties. American Society of Agronomy, Madison, Wisconsin.
Brokenshaw, D. W., Warren, D. M. and Werner, O. (eds.), (1980). Indigenous Knowledge Systems and Development. University Press of America, Lanham, Maryland.
Brush, S. B., Carney, H. J., and Huaman, Z. (1981). Dynamics of Andean potato agriculture.
Economic Botany 35: 70-88.
Bye, R. A., Jr. (1981). Quelites-ethnoecology of edible greens-past, present and future. Journal o f Ethnobiology 1: 109-123.
Byerlee, D., Collinson, M., Perrin, R., Winkelman, D., and Biggs, S. (1980). Planning Technologies Appropriate to Farmers-Concepts and Procedures. Centre International de
Mejoramiento de Maiz y Trigo (CIMMYT), Mexico City.
Chambers, R. (1983). Rural Development: Putting the Last First. Longman, New York.
Chang, J. H. (1977). Tropical agriculture: Crop diversity and crop yields. Economics and Geography 53: 241-254.
CIMMYT. (1974). The Puebla Project: Seven Years o f Experience: 1967-1973. Centro Internacional de Mejoramient de Maiz y Trigo, Mexico City.
Clawson, D. L. (1985). Harvest security and intraspecific diversity in traditional tropical agriculture. Economic Botany 39: 56-67.
Conway, F. R. (1985). Agricultural ecology and farming systems research. Paper presented at
the Farming Systems Workshop held at Hawkesbury Agricultural College, Australia.
Cruz, C. (1949). Estudio Agrological Regional del Estado de Tlaxcala, Mexico, D.F.
deJanvry, A. (1981). The Agrarian Question and Reformism in Latin America. The John
Hopkins University Press, Baltimore.
Denevan, W. M., Treacy, J. M., Alcorn, J. B., Padoch, C., Denslow, J., and Flores Paitan,
S. (1984). Indigenous agroforestry in the Peruvian Amazon: Bora indian management
of swidden fallows. Intereieneia 9: 346-357.
Donkin, R. A. (1979). Agricultural terracing in the aboriginal New World: Viking Fund. Publications in Anthropology 56: 1-196.
Egger, K. (1981). Ecofarming in the tropics - characteristics and potentialities Plant Research
and Development 13: 96-106.
Farrell, J. (1984). The role of trees in Tlaxcala mixed farming systems. Master's thesis, University of California, Berkeley.
Gliessman, S. R., Garcia, R., and Amador, M. A. (1981). The ecological basis for the application of traditional agricultural technology in the management of tropical agro-ecosystems.
Agro-Ecosystems 7: 173-185.

220

Altieri and Trujillo

Grossman, L. S. (1984). Feasants, Subsistence Ecology and Development in the Highlands of

Papua, New Guinea. Princetor~ University Press, New Jersey.
Harlan, J. R. (1976). Genetic 1~sources in wild relatives of crops. Crop Science 16: 329-333.
Harwood, R. R. (1979). Small Farm Development. 1ADS Development Oriented Literature Series.
Westview Press, Boulder, Colorado.
Klee, G. A. (ed.) (1980). World Systems of Traditional Resource Management. J. Wiley and
Sons, New York.
Letourneau, D. K. (1983). The effects of vegetational diversity on herbivorous insects and associated natural enemies: Examples from tropical and temperate agroecosystems. Ph.D.
dissertation, University of California, Berkeley.
Marten, G. G. (1986). Traditional Agriculture in Southeast Asia: A Human Ecology Perspective. Westview Press, Boulder, Colorado.
Moreno, L. E., Trujillo, J., and Ruiz, S. (1981). Reconocimiento de lasPlantas SilvestresAbundantes en Cultivos de Maiz y Frijol en Tlaxcala. Memorias de II Congreso Nacional de
la Ciencia de la Maleza, Chapingo, Mexico, pp. 61-67.
Nair, P. K. R. (1984). Soil Productivity Aspects of Agroforestry. Science and Practice of
Agroforestry. ICRAF, Nairobi, Kenya.
Oasa, E. K. (in press). Farming systems research: A change in form but not in content. Human
Organization.
Posey, D. (1983). Indigenous ecological knowledge and development of the Amazon. In Moran, E., (ed.), The Dilemma of Amazonian Development. Westview Press, Boulder,
Colorado.
Shaner, W. W., Phillipp, P. F., and Schnehl, W. R. (1982). Farming Systems Research and
Development: Guidelines for Developing Counwies. Westview Press, Boulder, Colorado.
Sprague, M. A., and Triplitt, G. B. (1986). No- Tillage and Surface- Tillage Agriculture. J. Wiley
and Sons, New York.
Toledo, V. M., Carabias, J., Mapes, C., and Toledo, C. (1985). Ecologia y Autosuficiencia
Alimentaria. Siglo Veintiuno Editores, Mexico.
Vermeer, D. E. (1979). The tradition of experimentation in swidden cultivation among the Tiv
of Nigeria. Applied Geography Conference 2: 244-257.
Weinstock, J. A. (1983). Rattan: Ecological balance in a Borneo rainforest swidden. Economic
Botany 37: 58-68.
Wilken, G. C. (1969). Drained-field agriculture: An intensive farming system in Tlaxcala, Mexico
Geographical Review 59: 215-241.
Wilken, G. C. (1970). The ecology of gathering in a Mexican farming region. Economic Botany
24(3): 286-295.
Wilken, G. C. (1977). Integrating forest and small-scale farm systems in middle America. AgroEcosystems 3: 291-301.
Williams, D. E. (1985). Tres arvenses Solanaceas comestibles y su proceso de domesticaci6n
en Tlaxcala, Mexico. Master's thesis, Colegio de Postproduodos, Chapiny~i, Mexico.
Zandstra, H. G., Price, E. C., Litsinger, J. A., and Morris, R. A. (1981). A Methodology for
on-Farm Cropping Systems Research. International Rice Research Institute, Los Banos,
Philippines.
Zinke, P. J. (1962). The patterns of influence of individual forest trees on soil properties. Ecology 43: 130-135.